Final Report on CIPM key comparison of 1 kg standards in stainless steel (CCM.M-K1)

Abstract

We report a key comparison of 1 kg stainless steel mass standards carried out between February 1995 and October 1997. Organized within the Consultative Committee for Mass and Related Quantities (CCM), the participants included 14 national metrology institutes (NMIs) and the BIPM, with the BIPM serving as pilot laboratory. The 14 NMIs were divided into two groups, each of which determined the mass of two cylindrical travelling standards. In preliminary measurements, the pilot laboratory had determined the important physical properties of the four travelling standards and had monitored their mass stability over time. Measurements within the two groups then proceeded in parallel, with the four travelling standards returning several times to the pilot laboratory in order to monitor their stability. One of the present authors (MG) shows in appendix 3 how to treat comparisons among participants in different 'loops'. As shown in appendix 4, the mass reported by each NMI was derived indirectly from the mass of its national prototype, introducing a significant correction for air buoyancy during this step.The median of all results was taken to be the reference value for this key comparison. The degrees of equivalence are shown in tables 3–5 and summarized graphically in figure 7.Main text. To reach the main text of this paper, click on Final Report. Note that this text is that which appears in Appendix B of the BIPM key comparison database kcdb.bipm.org/.The final report has been peer-reviewed and approved for publication by the CCM, according to the provisions of the Mutual Recognition Arrangement (MRA).

Mass Measurement of 1-kg Silicon Spheres for Determination of the Avogadro and Planck Constants

“…94), stainless steel (weights used in the CCM.M-K1 [21], T4, and weights used in the CCM.M-K4 [22]), and single-crystal silicon (AVO#3 [23], AVO28-S5 [2], and AVO28-S8 [2]) over a period from 1996 to 2011. Fig.…”

Establishing confidence from measurement comparisons

“…The consistency test against a weight-limited mean must include the consequent departures from being analytic chi-squareds, which can be evaluated by Monte Carlo simulation. In some cases, consistency testing against the simple mean has seemed attractive (Marullo-Reedtz et al 2003, Stock et al 2006, and in others testing against the median has been seen as overcoming some deficiencies in the general acceptance of the measurement results and their uncertainties (Aupetit et al 2004). Whatever algorithm is used to select a 'virtual right answer' from peer results, a chi-squared-like test can be applied to that KCRV and rigorously evaluated by Monte Carlo simulation of the claimed uncertainty distributions (Steele and Douglas 2005).…”

The Strategic Role of Inter-Laboratory Comparison Among International Institutions to Assure Confidence: Report of the INRIM and the European Spatial Agency Comparison on Electrical Quantities

“…To obtain the KCRV the measurements and uncertainties of all the participating Laboratories are considered, formally treating all the participating Laboratories at the same technical level. For this case, the rules to evaluate the results were given by Cox in [4] and applied for example in the comparisons [5][6][7]. Another kind of ILCs involves a NMI and secondary Laboratories of the same country, so belonging to the same MS.…”